A GPS-free Power Grid Monitoring System over
Mobile Platforms
Haoyang Lu, Lingwei Zhan Student Member, IEEE, Yilu Liu, Fellow, IEEE and Wei Gao, Member, IEEE
Department of Electrical Engineering and Computer Science
University of Tennessee
Knoxville, TN, USA
Email:{hlu9,lzhan,liu,weigao}@utk.edu
Abstract—In smart grids, the deviation of frequency and phase
angle serves as an important indicator of abnormal events, the
detection of which is critical to maintain reliable and stable
operations of the power grid. However, the high cost and low
accessibility of current power grid monitoring systems restrict
their large-scale deployment over highly distributed microgrids.
In this paper, we present a practical frequency and phase
angle monitoring system design over mobile platforms, which
significantly reduces the cost of power grid monitoring by
utilizing the functionalities of modern mobile devices. We harvest
the 200 Hz Primary Synchronization Signal (PSS) embedded
in the 4G Long Term Evolution (LTE) cellular signal for time
synchronization in power grid monitoring, replacing the PulsePer-Second (PPS) signal that is retrieved from GPS receivers
and being used by current monitoring systems. We implement
our system over realistic smartphones with a few peripheral
hardware, and realize a frequency monitoring accuracy of 0.2
mHz and a phase angle monitoring accuracy of 0.01 rad.
Experiment results compared with the traditional Frequency
Disturbance Recorders (FDRs) verify the effectiveness of our
proposed system.
Index Terms—Time synchonization, Long Term Evolution
(LTE), Smartphone
I. I NTRODUCTION
Power grid is a critical infrastructure of the modern society,
but is susceptible to various types of disturbances. When a
significant disturbance occurs, the frequency and the phase
angle of the power grid vary in both time and space, and,
in many ways, exhibit the characteristics of electromechanical
wave propagation phenomenon [1]. Therefore, the real-time
awareness of phasor state could serve as a significant indicator
of the power grid stability [2], and is indispensable for cybercontrolled smart grid construction.
How to obtain the phasor state more accurately and
efficiently has been an active research topic for decades.
Current power grid monitoring systems allow direct
measurement of frequency and voltage phase angle by
installing synchrophasors at either high-voltage transmission
level [3] or low-voltage distribution level [4], [5]. These power
grid monitoring systems, although having been proved to be
effective in wide-area power grid infrastructure, are generally
considered unsatisfactory for monitoring the operating status
of the newly emerging distributed power systems, socalled microgrids [6]. Microgrids enable local integration
This work was supported in part by the Engineering Research Center
Program of the National Science Foundation and the Department of Energy
under NSF Award Number EEC-1041877 and the CURENT Industry
Partnership Program. Pending provisional patent has been filed by the
University of Tennessee.
of energy generation, distribution, and storage at the
consumer level for better power system efficiency and control
of demand [7]. The decentralization of microgrids poses
higher requirements on the installation cost and accessibility
of current synchrophasors, which are too expensive and
inconvenient to be deployed into individual households in high
volume. For example, the installation cost of one FDR is
around $2,000, and it will cease to work when GPS signal
is unstable or unavailable for indoor scenarios. This high cost
and inconvenience would also reduce the end-users’ incentives
of having synchrophasors in their home energy systems.
In this paper, we present a practical system design
which bridges the gaps between current power grid
monitoring systems and the unique requirements of
microgrid monitoring, by fully unleashing the capabilities
of modern mobile platforms, particularly smartphones, in
computation, communication, and storage. We proposed
a novel time synchronization approach by harvesting the
Primary Synchronization Signal (PSS) in the 4G LTE cellular
radio, which significantly increases the system flexibility
by eliminating the requirement of GPS reception and line
of sight to the satellites. In addition, a low-cost monitoring
prototype is implemented on the smartphone platform with a
small quantity of peripheral hardware, which can be further
integrated into the smartphone charger for better flexibility
and convenience. The smartphone-based system lowers both
the manufacturing and installation costs of the monitoring
system. The accuracy of the frequency and phase angle
measurements are 0.2 mHz and 0.01 rad, respectively.
The rest of this paper is organized as follows. Section
II provides an overview of the system design. Section
III presents our time synchronization approach based on
harvesting the PSS signal from LTE radio. Section IV presents
our performance evaluation in comparison with FDRs. Section
V concludes the paper and discusses the future work.
II. OVERVIEW
In this section, we first briefly discuss our motivation of
developing the power grid monitoring system over mobile
platforms, particularly smartphones, and adopting the time
signal from LTE for time synchronization. Afterwards, we
present the hardware and software designs of our power grid
monitoring system.
A. Motivation
1) Motivation for using mobile platforms: Modern mobile
platforms, especially smartphones, are considered as ideal
AC Power
Smartphone
Voltage
Regulator
5V DC
Antenna
5,10,15V DC
Downconverting
Mixer
5V DC
Amplifier &
Filter
Voltage
data
Pulse Capture
& Processing
PSS Harvest Circuit
RESEND
(if nec)
External USB
Interface
START
Current
Transformer
Verizon 4G LTE coverage
Amplifier &
Low Pass filter
External
ADC
Microprocessor
Fig. 1. The coverage of Verizon 4G LTE in US
Voltage Transform Circuit
A/D Sampling Circuit
(a) System design
1 frame (10ms)
Antenna
1 subframe (1ms)
0
1
2
3
5
4
1 slot
0 1 2 3 4 5 6 0 1 2 3 4 5 6
1 OFDM
symbol
6
7
8
9
Voltage
Regulators
1 slot
0 1 2 3 4 5 6 0 1 2 3 4 5 6
:PSS
:SSS
Fig. 2. Synchronization signals in LTE FDD downlink
choices to reduce the manufacturing cost of the power
grid monitoring system. The essential components of a
synchrophasor – powerful computation units, Internet access
component and graphic interface component have all been
provided by the modern smartphone platform. In particular,
with the rapid development of IC and semiconductor
manufacturing, the cost of a smartphone has been significantly
reduced, and faster CPUs and more functionalities are
integrated into the smartphone platform. For example, the
Samsung Nexus S smartphone, with a price around 100
dollars, is equipped with 1 GHz ARM Cortex-A8 CPU, 512
MB RAM, GPS receiver (without analog PPS output), 4.0-inch
LCD screen, WiFi connection and cellular network access.
These capabilities make the modern smartphone platform be
an ideal choice for implementing the microgrid monitoring
systems.
2) Motivation for harvesting the time signal from 4G LTE
radio: A globally synchronized clock with high accuracy and
stability is required in the phase angle measurement. Simply
speaking, a 15 milliseconds timing error, which is usually
the upper bound of Network Timing Protocol (NTP) timing
error, corresponds to an unacceptable measurement error of
330 degrees in a 60 Hz system. The PPS signal provided by the
GPS system, with the accuracy in nanoseconds, are generally
used in current synchrophasors systems. However, the GPS
signal is usually unstable or unavailable in the indoor scenario
where microgrids are usually operated. Therefore, a pervasive
signal with the same level of accuracy is highly desired as an
alternative of GPS. eLoran timing method [8] was successfully
applied in synchrophasor measurement area and showed even
better accuracy than GPS. However, eLoran timing method
relies on high power transmitters, which results in high cost.
With its high data transmission rate and spectrum utilization
efficiency, the 4G Long Term Evolution (LTE) has gained
an extensive nationwide coverage, as shown in Fig. 1 [9],
and is being widely deployed worldwide. An enhanced base
station (eNodeB) of LTE is strictly synchronized with GPS
or the Precision Time Protocol (PTP) [10]. The cell ID in
MCU
Module
PSS Harvest
Circuit
Smartphone
(b) System implementation
Fig. 3. Power grid monitoring over mobile platforms
the LTE network is then defined within two synchronization
signals, namely Primary Synchronization Signal (PSS) and the
Secondary Synchronization Signal (SSS). Fig. 2 illustrates the
LTE frame format and the location of synchronization signals
under Frequency-Division Duplexing (FDD) mode. The PSS
signal repeats every 5 ms and therefore can be used as the
time synchronization signal.
B. Hardware design
Our developed power grid monitoring system consists
of voltage regulators, a voltage transform circuit, a
microprocessor-controlled (MCU-controlled) AD sampling
circuit, a PSS signal harvest module and an Android-based
smartphone. The system design and implementation are shown
in Fig. 3(a) and Fig. 3(b), respectively. The voltage regulator
part outputs the DC power to power up the smartphone,
MCU module and PSS harvest module. Current transformer
takes an analog voltage signal from 120V wall outlet and
transforms the AC power into the voltage range for analogto-digital conversion (ADC) purpose. An 8-bit microprocessor
ATmega328 (Arduino Uno board) is used to sample the voltage
waveform through external AD converter at the sampling
frequency of 1,440 Hz, and send the raw data to smartphone
every 100 ms for processing.
The communication between the MCU and the smartphone
is conducted by the USB host controller IC MAX3421E (USB
host shield) [11]. The MAX3421E host controller contains the
digital logic and analog circuitry necessary to implement a
full-speed host compliant to USB specification v2.0. Under
this circumstance, the smartphone behaves as USB slave in
relation to the USB host chip, and can communicate with
the MCU. The smartphone behaves as being connected to the
desktop computer and will be charged while being connected
to the USB host controller. The MCU communicates with the
USB host shield through serial peripheral interface (SPI) bus.
NTP
SERVER
FNET
SERVER
Frequency
Angle
Amplitude
NTP
Request
NTP
Response RESEND
(optional)
AD
Samples
...
USB
Communication
Component
Internet Access
Component
UTC TIME
PSS
(ZC sequence)
MCU
Frequency
Estimation
Component
...
0
1
...
...
...
4
5
6
34
35
DC
36
37
38
...
...
...
33
65
66
67
70
71
62 subcarriers (excluding DC)
72 subcarriers = 6 RB (1.08 MHz)
Sampled data
Flexible Bandwidth (1.4 to 20 MHz)
Fig. 5. Transmission of PSS signal in the frequency domain
Frequency Results
with Timestamps
Graphical
Interface
Component
Software Components
Application GUI Layout
LNA
Fig. 4. Software components in the Android application
LNA
BPF
ED
Clock
output
BPF
C. Software design
At the smartphone, the frequency estimation algorithm and
visualization programs are packaged into one Android
application. To keep the application responsive, the
functionality of voltage data reception and the expensive
computation of frequency estimation are distributed to an
independent application thread, instead of residing in the
application main thread or UI thread.
As shown in Fig. 4, our power grid monitoring application
consists of four major components, which are explained in
detail as follows:
1) USB communication component. Through the USB
connection, the smartphone will receive the sampled
measurement data from MCU. Our application will also
monitor the connection status of the USB accessory.
Once the smartphone is connected to the USB host
controller with a correct hardware signature, our
application will run automatically and build the USB
connection as “plug-and-play”.
2) Frequency estimation component. The phasor-based
algorithms [12] estimating the power grid frequency and
phase angles are implemented in this component.
3) Internet access component. Network access is
necessary to obtain the NTP timing information and
to upload the measurement results to the wide-area
monitoring network Frequency Monitoring Network
(FNET) operated by the University of Tennessee.
4) Graphic interface component. The information of
frequency, phase angle and AC voltage are displayed at
the smartphone for better user interaction. The graphic
interface component is placed in the main thread.
III. D ESIGN D ETAILS
A. Harvesting the Primary Synchronization Signal (PSS)
The LTE cellular networks are the most advanced cellular
communication infrastructure being deployed in the US, and
its availability is rapidly increasing throughout the nation.
In LTE networks, to achieve high data transmission rate,
Orthogonal Frequency Division Multiple Access (OFDMA) is
utilized as the physical layer technique in the downlink data
transmission. The cell ID can be calculated as:
Cell
1
2
NID
= 3NID
+ NID
1
NID
(1)
where
∈ 0, 1, ..., 167 is the cell identity group and is
2
located in SSS signal, and NID
∈ 0, 1, 2 is the cell identity and
VCO
Fig. 6. System block diagram of the PSS signal harvesting circuit
is located in PSS signal. The SSS signal indicates the frame
timing as they are different within a frame, while PSS signal
indicates the OFDM symbol timing as they are the same within
a frame. The sequence used for the PSS is generated from
a frequency domain Zadoff-Chu (ZC) sequence according to
[13]:
(
πun(n+1)
e−j 63
n = 0, 1, ..., 30
cu (n) =
(2)
πu(n+1)(n+2)
−j
63
e
n = 31, 32, ..., 61
where the ZC root u for the PSS sequence is 25,29,34 that
2
= 0,1,2, respectively.
corresponds to the value of NID
Comprised with three Zadoff-Chu sequences in frequency
domain, the PSS signal maps to the central 62 subcarriers
around DC (within the central six Resource Blocks (RBs)),
enabling the frequency mapping of the synchronization signals
to be invariant with respect to the system bandwidth, which
varies from 1.4 MHz to 20 MHz.
The frequency of PSS signal (200 Hz) is far less than the
bandwidth of data transmitted (in the order of 1 MHz). Since
our purpose of PSS detection is not to decode the signal
but only to identify the arrival of the PSS signals, the PSS
signal can be detected based on the scheme shown in Fig. 6.
A Voltage-Controlled Oscillator (VCO) is used to detect the
frequency band with the strongest signal strength. The signal
in 1900 MHz frequency band is selected and downconverted
to 200 kHz intermediate frequency (IF) output. The PSS signal
would be transformed as a pulse after passing the bandpass
filter with a bandwidth of 120 kHz and the envelope detector.
The MCU will capture the rising edge of the PSS pulses as
the trigger to start a new sampling cycle.
B. Coordination between smartphone and MCU
The major challenge of such coordination is that the local
clock of the MCU will continuously drift due to the dynamic
characteristics of the crystal oscillator, as well as various
environmental factors such as temperature. As a result, the
actual time period for each AC waveform sampling cycle
may not be accurately set as expected. For example, a time
period set to be the 100 ms by the MCU, which corresponds
to the reporting rate in FNET, may be actually 99.8 ms
or 100.1 ms due to the local clock drift. Such inaccurate
1 sampling cycle (0.1 s)
PSS Harvest
START Signal
AC POWER
MCU (ADC)
142 143
0
1
……
142 143
0
1
Sample Index
TIME
FREQUENCY
PHASE ANGLE
AMPLITUDE
Data
Frequency
correction estimation
Data
Frequency
correction estimation
14:31:12.1
60.0147 Hz
5.3321 rad
120.2 V
FNET Server
2.3395
60.0002
2.3295
60
59.9998
14:31:12.2
60.0141 Hz
5.3317 rad
120.1 V
2.3195
2.3095
2.2995
59.9996
59.9992
RESEND Signal
(if necessary)
Smartphone
2.3495
60.0004
2.2895
59.9994
……
Waveform
Samples
60.0006
Angle (rad)
Sampling Cycle
START Signal
Frequency (Hz)
PSS Harvest
Window
PSS Harvest
Circuit
2.2795
0
500
1000
1500
Time (0.1s)
2000
2500
0
500
(a) Frequency
1000
1500
Time (0.1s)
2000
2500
(b) Phase angle
Fig. 9. Frequency and phase angle monitoring over 60 Hz Doble
output
Data publish
35
Smartphone−based prototype
Frequency Disturbance Recorder
60.035
Smartphone−based prototype
Frequency Disturbance Recorder
60.03
angle (rad)
Fig. 7. System coordination between the smartphone and MCU
frequency (Hz)
30
60.025
60.02
25
20
15
10
60.015
5
60.01
500
1000
1500
time (0.1s)
2000
2500
500
(a) Frequency
1000
1500
time (0.1s)
2000
2500
(b) Phase angle
Fig. 10. Frequency and phase angle monitoring over 120V wall
power
Fig. 8. Experiment setup
sampling frequency will result in the residue problem [14],
i.e., the position of the first sample in one sampling cycle is
different from that in another cycle, and the residue could be
accumulated over time.
Our proposed method of determining the starting time of a
new sampling cycle is shown in Fig. 7. Our monitoring system
starts a new sampling cycle on every 100 ms, which is to
be compatible with the reporting rate of the FNET network,
and uses 1,440 Hz as the sampling frequency to ensure the
measurement accuracy (12 or more data samples per period
is needed in a 60 Hz power system). Correspondingly, 144
data samples are recorded in each sampling cycle. Our basic
approach to addressing the above residue problem is to enable
the PSS pulse detection after sampling the 142-nd samples
(98.6ms) in current sampling cycle. The pulses received in
this period cycle is assumed to be the PSS signal first, and
the interval between the pulse and the last effective pulse is
calculated. If the interval ∆t < ε where ε is the threshold,
then the pulse is considered as an effective PSS output. The
detection window will be closed and a new sampling cycle
starts immediately. Otherwise, the pulse is assumed to be noise
and if no effective pulses are detected in the detection window,
the MCU will start a new sampling cycle at the time when
sampling the 144th samples.
IV. P ERFORMANCE E VALUATION
To evaluate the accuracy and effectiveness of the power
grid monitoring of our proposed system, we test our system
against the traditional FDR device. The system setup is shown
in Fig. 8. The Doble F6150 power system simulator with
GPS satellite synchronization is used to generate the standard
AC power. The frequency and the phase angle accuracy of
Doble F6150 simulator are 0.5 Part-Per-Million (PPM) and 0.1
degree respectively. Both standard power generator and regular
AC wall power are used in our experiments. In our prototype,
the measurement results are written to a local textfile for
review, while we extract the measurement results of FDR
through FNET server.
A. Frequency and Angle measurement
First, the frequency measurement result over standard 60
Hz waveform generated by Doble is shown in Fig. 9. The
error of the frequency measurement of our system is 0.2
mHz, compared with 0.22 mHz of the FDR device. For angle
measurement, the standard deviation is used to evaluate the
error. In the ideal condition, as to the waveform with fixed
frequency, if 10 Hz reporting rate is applied, the curve of phase
angle would be a straight line with fixed slope, in particular, a
horizontal line under 60Hz condition. The standard deviation
of the angle measurement result is 0.01 rad, which corresponds
to 32us in 60 Hz system, compared to 0.002 rad of the FDR
device. Second, the frequency measurement results over the
120V AC wall outlet are shown in Fig. 10. Our system is able
to efficiently capture the trends of frequency deviations over
time. Meanwhile, the difference between our system and the
FDR measurements is 0.4 mHz. For angle measurement, since
the PSS signal and the GPS signal are not precisely aligned,
there is a nearly constant drift around the results. If eliminate
the DC offset of two results, the difference between two curves
is 0.011 rad.
In comparison with FDR, an 16-bit ADC used in our system
lowers the quantization distortion in the AD process which
leads to the increase of the frequency monitoring accuracy of
the prototype. Angle measurement has a more strict demand
on the clock accuracy than frequency measurement. Currently,
the angle monitoring accuracy of our system is lower than that
of FDR due to a couple of reasons. First, the timing error is
introduced when capturing the PSS signal. The PSS signals
are harvested in the form of pulses, and then the rising edge
12
error of phase angle
SNR
0.015
0.01
10
SNR (dB)
Error of phase angle (rad)
0.02
0.005
0
uncovered
partially covered
totally covered
8
Measurement condition
Fig. 11. Accuracy of phase angle measurement under different SNR
of PSS signal
of the pulses will be captured by the MCU through external
interrupt. Compared to the PPS pulses with a duration of 1.2
us and a steep rising edge, the slope of the rising edge of
the PSS pulse is flatter and can be affected by the strength
of the signal. Second, timing error exists in the timestamps
compared to the real UTC time (GPS signal). That is, the
time point of a frequency measurement is not exactly aligned
with that of the GPS-based FDR. Different from GPS signal,
there may be no PSS signal in the boundary of every two UTC
seconds. Third, due to the involvement of both inductive and
capacitive components such as transformers and analog filters
with different time constants, possible phase lag or advance
may be introduced into the frequency monitoring. This may
induce a constant time deviance between measurements of our
system and the FDR.
B. Measurement Accuracy vs. Signal-to-Noise Ratio
To evaluate the relationship between the measurement
accuracy and the strength of the LTE cellular signal reception,
we emulate the environment with different signal-to-noise ratio
(SNR) by using aluminium foil to cover the antenna of the
PSS harvest antenna. The complete coverage of the foil onto
the antenna corresponds to the worst SNR environment. The
performance of the phase angle measurement with respect to
different SNR is depicted in Fig. 11. In Fig. 11, the PSS
harvest module can achieve the highest SNR of 11.49 dB
under no cover with the error of 0.01 rad on phase angle
monitoring. As the coverage area of the foil increases, the
error increases to 0.0127 and 0.015 rad in partially covered
and totally covered situation, respectively. The reason of such
a degradation partially results from the method of capturing the
PSS pulses. As the noise level increases, the voltage amplitude
of the PSS signal become closer to that of the noises, which
refers to the slope of the rising edge of the PSS signal is small
and susceptible to the noise. Since we regard the incoming
of the rising edge of the PSS signal as the synchronization
signal, the physical property of the MCU will introduce more
uncertainty in a low SNR condition.
V. C ONCLUSION AND F UTURE W ORK
In this paper, we presented the design and implementation
of a GPS-free smartphone-based power grid frequency and
angle monitoring system. In this system, we proposed to use
the Primary Synchronization Signal in the 4G LTE cellular
network as the synchronization source as an alternative of
GPS signal. The extendability of the smartphone platform
enables more functionalities to be further integrated into
the system. The experiments compared with FDR devices
verify the effectiveness of the system on frequency and angle
monitoring. Our future work will focus on further improving
the performance of PSS harvest module, simplifying the power
grid monitoring system, and integrating more functionalities
of power grid monitoring onto mobile platforms. First, we
will modify the method in capturing the PSS signal by MCU.
Instead of capturing the rising edge of the pulse, which is
constrained by the property of the MCU circuitry and of more
uncertainty in a low SNR condition, a high-speed ADC will
be used to collect the waveform of the IF signal, and the peak
detection algorithms would be applied to the IF signal to detect
the peak of PSS pulse as the start of a sampling cycle. Second,
we will fabricate all the external components including MCU
board, PSS harvest module and the voltage transform circuit
into one PCB board, so that our monitoring system could be
integrated into the smartphone charger for better accessibility.
Third, our system will be further extended to incorporate
more functionalities such as power quality measurement that
monitors power grid operations in a much wider range of
frequency spectrum.
R EFERENCES
[1] J. S. Thorp, C. E. Seyler, and A. G. Phadke, “Electromechanical wave
propagation in large electric power systems,” Circuits and Systems I:
Fundamental Theory and Applications, IEEE Transactions on, vol. 45,
no. 6, pp. 614–622, 1998.
[2] H. Karimi, M. Karimi-Ghartemani, and M. R. Iravani, “Estimation of
frequency and its rate of change for applications in power systems,”
IEEE Transactions on Power Delivery, vol. 19, no. 2, pp. 472–480,
2004.
[3] V. Venkatasubramanian, H. Schattler, and J. Zaborszky, “Fast timevarying phasor analysis in the balanced three-phase large electric power
system,” IEEE Transactions on Automatic Control, vol. 40, no. 11, pp.
1975–1982, 1995.
[4] Y. Zhang, P. Markham, T. Xia, L. Chen, Y. Ye, Z. Wu, Z. Yuan, L. Wang,
J. Bank, J. Burgett et al., “Wide-area frequency monitoring network
(FNET) architecture and applications,” IEEE Transactions on Smart
Grid, vol. 1, no. 2, pp. 159–167, 2010.
[5] L. Zhan and Y. Liu, “Improved wls-tf algorithm for dynamic
synchronized angle and frequency estimation,” in PES General Meeting
— Conference Exposition, 2014 IEEE, July 2014, pp. 1–5.
[6] C. W. Gellings, M. Samotyj, and B. Howe, “The future’s smart delivery
system,” IEEE Power and Energy Magazine, vol. 2, no. 5, pp. 40–48,
2004.
[7] F. Giraud and Z. M. Salameh, “Steady-state performance of a gridconnected rooftop hybrid wind-photovoltaic power system with battery
storage,” IEEE Transactions on Energy Conversion, vol. 16, no. 1, pp.
1–7, 2001.
[8] L. Zhan, D. Zhou, T. King, Y. Liu, E. Johannessen, J. Alexander, and
B. Boza, “Improvement of timing reliability and data transfer security
of synchrophasor measurements,” in T D Conference and Exposition,
2014 IEEE PES, April 2014, pp. 1–5.
[9] Verizon Wireless, “LTE network coverage 2014,” http://www.
verizonwireless.com/wcms/consumer/4g-lte.html.
[10] “IEEE standard for a precision clock synchronization protocol for
networked measurement and control systems,” IEEE Std 1588-2008
(Revision of IEEE Std 1588-2002), pp. c1–269, July 2008.
[11] MAXIM Integrated (2007), “MAX3421E, usb peripheral/host controller
with spi interface,” copyright Maxim Integrated Products.
[12] A. G. Phadke, J. S. Thorp, and M. G. Adamiak, “A new measurement
technique for tracking voltage phasors, local system frequency, and rate
of change of frequency,” IEEE Transactions on Power Apparatus and
Systems, no. 5, pp. 1025–1038, 1983.
[13] 3rd Generation Partnership Project (3GPP); Technical Specification
Group Radio Access Network, “Evolved universal terrestrial radio access
(E-UTRA); physical channels and modulation (release 10) TS 36.211
(version 10.0.0),” Tech. Rep., 2010.
[14] J. Zuo, “The frequency monitor network (FNET) design and
situation awareness algorithm development,” Ph.D. dissertation, Virginia
Polytechnic Institute and State University, 2008.